The temporal and spatial distribution (pharmacokinetics) of agents administered into the feline brain by direct infusion are being studied. Reasons for direct infusion are avoidance of the blood-brain barrier, decreased systemic exposure, and avoidance of systemic metabolic and immunologic interaction with the agent. The aim of the project is to develop methods to administer agents into the brain by direct infusion, and to achieve a homogeneous distribution throughout the volume of distribution. Agents for study are analogous to immunotoxins, 1-125 labeled iododeoxyuridine, and commercially available antineoplastic agents.